Fission Near 198Pb with AT-TPC at FRIB: Insights from Curtis Hunt
Curtis Hunt
S
t
u
d
y
i
n
g
F
i
s
s
i
o
n
n
e
a
r
1
9
8
P
b
w
i
t
h
A
T
-
T
P
C
a
t
F
R
I
B
Fission in AT-TPC
•
Interest in Fission
•
Method
Setup
•
AT-TPC
•
HEavy ISotope Tagger (HEIST)
Challenges
•
Finding Window and Pad Plane in Time
•
Identifying Fission Events
»
Algorithmic Method
»
Machine Learning Method
•
Space Charge
•
Signal Processing
Preliminary Results
Summary
O
u
t
l
i
n
e
Curtis Hunt, TPC2023 Workshop
, Slide 2
Probe of nuclear structure
•
Unexpected region of
asymmetry in mass yields
near
180
Hg
•
Our interest is in the
transition region near
198
Pb
Probe of low-density
equation of state
•
Compare barrier and mass
distributions to different
effective interactions
F
i
s
s
i
o
n
I
n
A
T
-
T
P
C
Curtis Hunt, TPC2023 Workshop
, Slide 3
M
o
l
l
e
r
P
a
n
d
R
a
n
d
r
u
p
J
2
0
1
5
P
h
y
s
.
R
e
v
.
C
9
1
0
4
4
3
1
6
Goal is to measure fission
properties of short-lived isotopes
•
Fission barrier, cross-section,
asymmetry of fission fragments
Nucleus excited through light ion
induced fusion-fission -> low angular
momentum
•
Fission of compound nucleus
Using an Active Target TPC
provides several advantages to the
study of fission using fusion-fission
F
i
s
s
i
o
n
I
n
A
T
-
T
P
C
Curtis Hunt, TPC2023 Workshop
, Slide 4
M
o
l
l
e
r
P
e
t
.
A
l
.
J
2
0
1
5
P
h
y
s
.
R
e
v
.
C
9
1
0
2
4
3
1
0
Using an Active Target allows for probing a large range of excitation
energies
•
Allows for measuring cross section over range of energies with single beam
Active Target allows for accurate determination of reaction vertex and
reconstruction of the event
•
Vertex used to obtain excitation energy and the angle between fission
fragments
Inverse kinematics allows for the use of radioactive isotopes in study of
fusion-fission reactions
Inverse kinematics boosts fragments forward
•
F
i
s
s
i
o
n
e
v
e
n
t
s
f
o
r
m
a
c
h
a
r
a
c
t
e
r
i
s
t
i
c
Y
p
a
t
t
e
r
n
t
o
w
a
r
d
t
h
e
p
a
d
p
l
a
n
e
F
i
s
s
i
o
n
I
n
A
T
-
T
P
C
Curtis Hunt, TPC2023 Workshop
, Slide 5
A
T
-
T
P
C
Curtis Hunt, TPC2023 Workshop
, Slide 6
AT-TPC (Active Target Time
Projection Chamber)
•
Reactions occur inside the gas
volume of the TPC with the gas
1000 mm long x 250 mm radius
cylindrical active volume
10240 micromegas pads
•
Perpendicular to beam direction
•
Triangular pads
•
Highly segmented inner region
Cylindrical design allows for use in
magnetic field
•
We did
not
use a magnetic field for
fission study
A
T
-
T
P
C
-
G
E
T
Curtis Hunt, TPC2023 Workshop
, Slide 7
Generic Electronics for TPCs (GET)
•
AGET chips
»
64 data channels + 4 FPN channels
•
Fixed Pattern Noise (FPN) channels
measure the baseline due to electronics
»
Preamplification and Shaping
•
AsAd (ASIC and ADC) boards
»
4 AGET chips per board
»
Digitizes signal from the AGET
•
CoBo (Concentration Board) Module
»
Up to 4 AsAd per CoBo
•
MuTAnT (Multiplicity, Trigger and
Time)
»
Manages the trigger for the connected
CoBos
E
.
C
.
P
o
l
l
a
c
c
o
e
t
a
l
.
N
I
M
A
8
8
7
(
2
0
1
8
)
8
1
-
9
3
I
d
e
n
t
i
f
y
i
n
g
R
a
r
e
I
s
o
t
o
p
e
B
e
a
m
s
Curtis Hunt, TPC2023 Workshop
, Slide 8
Simulated PID plot
Z: Element number
A: Mass number
q: Charge-state
H
E
a
v
y
I
S
o
t
o
p
e
T
a
g
g
e
r
(
H
E
I
S
T
)
Curtis Hunt, TPC2023 Workshop
, Slide 9
ToF
Adam Anthony
MSU Grad Student
E
x
p
e
r
i
m
e
n
t
a
l
S
e
t
u
p
Curtis Hunt, TPC2023 Workshop
, Slide 10
PID system (HEIST)
•
MCP and ion chamber
•
Si and HPGe support system
Degrader system
•
Fe degrader and ion chamber
Active target (AT-TPC)
•
4
He gas target
•
E* from 65 MeV down to 30 MeV
•
Trigger on MCP2 and pad plane coincidence
•
Identify pads showing energy deposition near window and pad plane
•
Fit leading edge of pad traces
•
W
i
n
d
o
w
:
t
0
g
i
v
e
s
t
i
m
e
e
l
e
c
t
r
o
n
s
d
r
i
f
t
e
d
f
r
o
m
w
i
n
d
o
w
r
e
a
c
h
p
a
d
p
l
a
n
e
•
P
l
a
n
e
:
t
0
g
i
v
e
s
t
i
m
e
f
i
s
s
i
o
n
f
r
a
g
m
e
n
t
h
i
t
s
p
a
d
p
l
a
n
e
F
i
n
d
i
n
g
W
i
n
d
o
w
/
P
a
d
P
l
a
n
e
T
i
m
e
Curtis Hunt, TPC2023 Workshop
, Slide 11
Bergen Kendziorski
MSU Undergraduate
Pad Plane
Window
Low gain region in center of TPC should omit beam events
In practice, we still record orders of magnitude more beam than fission
•
~ 2-3% of events are fission events
Algorithm was developed to tag as many unwanted events as possible
•
Set of ~2500 hand labeled events was used for development
•
Parameters that can be quickly & automatically calculated were found
»
P
o
l
a
r
a
n
g
l
e
s
o
f
s
t
r
a
i
g
h
t
-
l
i
n
e
t
r
a
c
k
s
,
s
p
r
e
a
d
o
f
e
-
h
i
t
s
r
e
c
o
r
d
e
d
i
n
x
-
y
p
l
a
n
e
,
e
t
c
.
•
Events are either labeled fission, beam, or “other” if parameters are
ambiguous
S
o
r
t
i
n
g
F
i
s
s
i
o
n
F
r
o
m
B
e
a
m
Curtis Hunt, TPC2023 Workshop
, Slide 12
Fission
Beam
Other?
S
o
r
t
i
n
g
F
i
s
s
i
o
n
F
r
o
m
B
e
a
m
:
R
e
s
u
l
t
s
Run: 206
Beam Rate: 3809
G
a
s
:
C
F
4
@
1
5
0
T
o
r
r
Run: 261
Beam Rate: 2136
G
a
s
:
C
F
4
@
2
0
0
T
o
r
r
“Other”:
25 Fissions
44 Beams
“Other”:
27 Fissions
111 Beams
Curtis Hunt, TPC2023 Workshop
, Slide 13
Events are sorted into “beam”,
“fission”, and “other” categories.
~2-3% of all events are sorted into
the “other” category
~15-30% of fission events are
sorted into “other”
Combining “fission” and “other”
categories yield 100% fission event
capture with ~30-40% events being
beam events
Joe Wieske
MSU Grad Student
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
A
p
p
r
o
a
c
h
T
o
C
l
a
s
s
i
f
y
i
n
g
F
i
s
s
i
o
n
a
n
d
B
e
a
m
E
v
e
n
t
s
Curtis Hunt, TPC2023 Workshop
, Slide 14
Confusion matrix from testing on hand-labeled data from the
other category
Use a machine learning model to
classify these “Other” events.
•
Worked
with Michelle Kuchera and
Raghu Ramanujan (Davidson College)
Train the model on labeled Fission
and Beam events from the
algorithmic approach
Apply the trained model to a small
subset of hand labeled “Other”
events for testing
•
81% Fission events in “Other”
correctly identified
•
~94-97% Fission events
captured with
algorithmic + ML methods combined
Poulomi Dey
MSU Undergraduate
Beam
Fission
Accumulated positive ions (
He
+
) drift
much slower than e
-
because of their
significantly lower mobility
•
Creates a line of positive charge along the
beam path
Some fission fragment tracks curve
inward
•
D
r
i
f
t
i
n
g
e
-
a
r
e
p
u
l
l
e
d
i
n
w
a
r
d
s
b
y
t
h
e
r
e
s
u
l
t
i
n
g
i
o
n
t
r
a
n
s
v
e
r
s
e
E
-
f
i
e
l
d
•
Effect is rate dependent
Results in wider folding angles and
reaction vertices offset downstream
•
Vertex location gives the excitation energy
•
Fragment directions give mass split
S
p
a
c
e
C
h
a
r
g
e
:
T
h
e
P
r
o
b
l
e
m
Curtis Hunt, TPC2023 Workshop
, Slide 15
Fission events are fit with “Y-track”
to extract observables (vertex,
angles)
For certain events we can
determine the vertex separately
from Y-track
•
The lighter fission fragments loose
much less energy over distance
than the heavier beam, creating a
discontinuity in dE/dx
These provide a metric to quantify
how well corrections are tuned
•
Proper correction will result in vertex
agreement between the two
methods
S
p
a
c
e
C
h
a
r
g
e
:
H
o
w
T
o
C
o
r
r
e
c
t
D
i
s
t
o
r
t
i
o
n
s
*
*
*
B
e
f
o
r
e
c
o
r
r
e
c
t
i
o
n
*
*
*
Curtis Hunt, TPC2023 Workshop
, Slide 16
Beam region dE/dx
Distortions to e
-
drift path can be modeled by solving Langevin
Equation
•
Electric field is a combination of the known TPC drift field and the unknown
space charge distribution
•
Space charge should build up more strongly towards entrance window
Input a model for the space charge distribution, then solve for drift
corrections
S
p
a
c
e
C
h
a
r
g
e
:
H
o
w
T
o
C
o
r
r
e
c
t
D
i
s
t
o
r
t
i
o
n
s
Curtis Hunt, TPC2023 Workshop
, Slide 17
S
p
a
c
e
C
h
a
r
g
e
:
C
o
r
r
e
c
t
i
o
n
R
e
s
u
l
t
s
RED
= Before
correction
BLUE
= After
correction
*
*
*
E
x
a
m
p
l
e
C
o
r
r
e
c
t
e
d
E
v
e
n
t
*
*
*
Curtis Hunt, TPC2023 Workshop
, Slide 18
Y-Fit Vertex (mm)
Y-Fit Vertex (mm)
ADC Vertex (mm)
Vertex Difference
- Before -
Vertex Difference
- After -
Joe Wieske
MSU Grad Student
When the charge distribution is short relative to the shaper rise time,
the total charge is measured by the shaped pulse height
When the distribution is long the pulse height does not represent the
total charge
•
Fission fragments move toward the pad plane, spending notable distance
over a given pad
D
e
c
o
n
v
o
l
u
t
i
o
n
–
S
h
a
p
e
r
E
f
f
e
c
t
s
Curtis Hunt, TPC2023 Workshop
, Slide 19
Input Charge
Output Signal
D
e
c
o
n
v
o
l
u
t
i
o
n
–
R
e
c
o
n
s
t
r
u
c
t
i
o
n
P
r
i
n
c
i
p
l
e
Curtis Hunt, TPC2023 Workshop
, Slide 20
J
.
G
i
o
v
i
n
a
z
z
o
e
t
a
l
.
N
I
M
A
8
4
0
(
2
0
1
6
)
1
5
-
2
7
650 mV internal AsAd pulser test used to get response function
•
Pulser signal is recorded by the FPN channels
Deconvolution of the pulser signal did not work due to very few time
buckets involved in the signal
•
Pulser signal is effectively a delta function
Shaped pulser signal was used as the response function
D
e
c
o
n
v
o
l
u
t
i
o
n
–
R
e
s
p
o
n
s
e
F
u
n
c
t
i
o
n
Curtis Hunt, TPC2023 Workshop
, Slide 21
Pulser Signal
[
i
in
(
t
)]
Shaped Pulser Signal
[
s
out
(
t
)]
The inverted signal messes up the return to zero
Add the positive signal to the negative signal to cancel it out
Some “glitches” needed to be removed and extrapolated
Shifted response function to start at 0 and normalized height to 1
D
e
c
o
n
v
o
l
u
t
i
o
n
–
R
e
s
p
o
n
s
e
F
u
n
c
t
i
o
n
Curtis Hunt, TPC2023 Workshop
, Slide 22
Response Function
[
h
(
t
)]
Shaped Pulser Signal
[
s
out
(
t
)]
Deconviolution corrects shape for undershoot due to response function
Gives reliable dQ/dt results for fission fragments
D
e
c
o
n
v
o
l
u
t
i
o
n
–
R
e
s
u
l
t
s
Curtis Hunt, TPC2023 Workshop
, Slide 23
Shaped Signal [
s
out
(
t
)]
Reconstructed Charge [
i
in
(
t
)]
P
r
e
l
i
m
i
n
a
r
y
R
e
s
u
l
t
s
Curtis Hunt, TPC2023 Workshop
, Slide 24
-
Fragment A
-
Fragment B
Time Buckets
Charge
Adam Anthony
MSU Grad Student
Missing pads
P
r
e
l
i
m
i
n
a
r
y
R
e
s
u
l
t
s
Curtis Hunt, TPC2023 Workshop
, Slide 25
-
Fragment A
-
Fragment B
Time Buckets
Charge
Adam Anthony
MSU Grad Student
Missing pads
Fission in AT-TPC
•
Experiment looking at fission near
198
Pb has been performed with AT-TPC
Challenges Solved
•
Beam particle identification
»
HEavy ISotope Tagger (HEIST)
•
Identifying Fission Events
»
Use a combination of algorithmic and machine learning methods
•
Space Charge
»
Solved Longevin equation to correct fission tracks
•
Signal Processing
»
Deconvolution for accurate dQ/dt (related to dE/dx)
Preliminary fission fragment asymmetry measurements
•
More results, including cross sections, to come
Future
•
Use rare isotope beams from FRIB to look at fission of more exotic isotopes
S
u
m
m
a
r
y
Curtis Hunt, TPC2023 Workshop
, Slide 26
Questions?
Q
u
e
s
t
i
o
n
s
?
Curtis Hunt, TPC2023 Workshop
, Slide 27
B
a
c
k
u
p
S
l
i
d
e
s
Curtis Hunt, TPC2023 Workshop
, Slide 28
S
o
r
t
i
n
g
F
i
s
s
i
o
n
F
r
o
m
B
e
a
m
:
E
x
a
m
p
l
e
P
a
r
a
m
e
t
e
r
Curtis Hunt, TPC2023 Workshop
, Slide 29
S
o
r
t
i
n
g
F
i
s
s
i
o
n
f
r
o
m
B
e
a
m
:
R
e
s
u
l
t
s
, Slide 30
Curtis Hunt, TPC2023 Workshop
Run: 210
Beam Rate: 3335
G
a
s
:
C
F
4
@
1
5
0
T
o
r
r
Trigger: DS Singles +
coinc
Run: 206
Beam Rate: 3809
G
a
s
:
C
F
4
@
1
5
0
T
o
r
r
Trigger: DS Singles +
coinc
Run: 211
Beam Rate: 6467
G
a
s
:
C
F
4
@
1
5
0
T
o
r
r
Trigger: DS Singles +
coinc
Run: 260
Beam Rate: 2106
G
a
s
:
C
F
4
@
2
0
0
T
o
r
r
Trigger: DS Singles
(5000) + coinc
Run: 261
Beam Rate: 2136
G
a
s
:
C
F
4
@
2
0
0
T
o
r
r
Trigger: DS Singles
(500) + coinc
Run: 271
Beam Rate: 7665
G
a
s
:
C
F
4
@
2
0
0
T
o
r
r
Trigger: DS Singles
(500) + coinc
“Other” = 25
Fissions
44 Beams
“Other” = 27
Fissions
111 Beams
D
e
c
o
n
v
o
l
u
t
i
o
n
–
R
e
c
o
n
s
t
r
u
c
t
i
o
n
P
r
i
n
c
i
p
l
e
Curtis Hunt, TPC2023 Workshop
, Slide 31
Pulser tests with internal AsAd pulser were used to obtain response
function of the GET preamp and shaper
0 V test to get backgrounds
•
Channel dependent baseline
•
Phase effect from circular buffer
Baselines usually corrected by FPN channels, but in pulser test FPN
channels have the pulser signal
D
e
c
o
n
v
o
l
u
t
i
o
n
–
B
a
s
e
l
i
n
e
C
o
r
r
e
c
t
i
o
n
s
Curtis Hunt, TPC2023 Workshop
, Slide 32
For long signals (such as for the beam) the deconvolution does not
reconstruct charge perfectly
I
t
e
r
a
t
i
v
e
D
e
c
o
n
v
o
l
u
t
i
o
n
Curtis Hunt, TPC2023 Workshop
, Slide 33
I
t
e
r
a
t
i
v
e
D
e
c
o
n
v
o
l
u
t
i
o
n
Curtis Hunt, TPC2023 Workshop
, Slide 34
Signal can be reconstructed by convoluting the charge with the
response function using the summation method
The difference between the source signal and the reconstructed signal
is not 0
The difference between the actual signal and the signal from the
reconstructed charge can be iteratively used to correct the charge.
•
1) Use the deconvolution method on the signal difference to obtain the
difference in “charge space” (go from “signal space” to “charge space”)
•
2) Subtract the result from the charge and obtain a new reconstructed signal
•
3) Calculate the difference between the source signal and new reconstructed
signal
•
4) Repeat
I
t
e
r
a
t
i
v
e
D
e
c
o
n
v
o
l
u
t
i
o
n
Curtis Hunt, TPC2023 Workshop
, Slide 35
4 iterations appears to be sufficient to obtain a relatively flat difference
I
t
e
r
a
t
i
v
e
D
e
c
o
n
v
o
l
u
t
i
o
n
Curtis Hunt, TPC2023 Workshop
, Slide 36
Cross-section
•
Energy determined from vertex location
»
Space Charge corrections crucial
•
Determination of Fission vs Beam
»
Standard + ML methods
O
b
s
e
r
v
a
b
l
e
s
f
r
o
m
A
T
-
T
P
C
Curtis Hunt, TPC2023 Workshop
, Slide 37
O
b
s
e
r
v
a
b
l
e
s
f
r
o
m
A
T
-
T
P
C
Curtis Hunt, TPC2023 Workshop
, Slide 38
S
i
m
u
l
a
t
i
o
n
o
f
f
i
s
s
i
o
n
e
v
e
n
t
s
Curtis Hunt, TPC2023 Workshop
, Slide 39
Simulated fission events
Experimental fission events
(all beam species)
Z ~ 32
Z ~ 52
Delve into the intriguing world of fission studies near 198Pb using the AT-TPC at FRIB. Supported by the DOE Office of Science, this research probes nuclear structure, fission properties, and fusion-fission reactions. By employing innovative techniques like the Heavy Isotope Tagger and active target TPC, researchers aim to measure fission properties of short-lived isotopes and uncover new insights into nuclear physics.
Download Presentation
Please find below an Image/Link to download the presentation.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. Download presentation by click this link. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.
E N D
Presentation Transcript
Studying Fission near 198Pb with AT-TPC at FRIB Curtis Hunt This material is based upon work supported by the U.S. Department of Energy Office of Science under Cooperative Agreement DE-SC0000661, the State of Michigan and Michigan State University. Michigan State University designs and establishes FRIB as a DOE Office of Science National User Facility in support of the mission of the Office of Nuclear Physics.
Outline Fission in AT-TPC Interest in Fission Method Setup AT-TPC HEavy ISotope Tagger (HEIST) Challenges Finding Window and Pad Plane in Time Identifying Fission Events Algorithmic Method Machine Learning Method Space Charge Signal Processing Preliminary Results Summary Curtis Hunt, TPC2023 Workshop , Slide 2
Fission In AT-TPC Probe of nuclear structure Unexpected region of asymmetry in mass yields near 180Hg Our interest is in the transition region near 198Pb Probe of low-density equation of state Compare barrier and mass distributions to different effective interactions Z N Moller P and Randrup J 2015 Phys. Rev. C 91 044316 Curtis Hunt, TPC2023 Workshop , Slide 3
Fission In AT-TPC Goal is to measure fission properties of short-lived isotopes Fission barrier, cross-section, asymmetry of fission fragments Nucleus excited through light ion induced fusion-fission -> low angular momentum Fission of compound nucleus Using an Active Target TPC provides several advantages to the study of fission using fusion-fission Moller P et. Al. J 2015 Phys. Rev. C 91 024310 Curtis Hunt, TPC2023 Workshop , Slide 4
Fission In AT-TPC Using an Active Target allows for probing a large range of excitation energies Allows for measuring cross section over range of energies with single beam Active Target allows for accurate determination of reaction vertex and reconstruction of the event Vertex used to obtain excitation energy and the angle between fission fragments Inverse kinematics allows for the use of radioactive isotopes in study of fusion-fission reactions Inverse kinematics boosts fragments forward Fission events form a characteristic Y pattern toward the pad plane Curtis Hunt, TPC2023 Workshop , Slide 5
AT-TPC AT-TPC (Active Target Time Projection Chamber) Reactions occur inside the gas volume of the TPC with the gas 1000 mm long x 250 mm radius cylindrical active volume 10240 micromegas pads Perpendicular to beam direction Triangular pads Highly segmented inner region Cylindrical design allows for use in magnetic field We did not use a magnetic field for fission study Curtis Hunt, TPC2023 Workshop , Slide 6
AT-TPC - GET Generic Electronics for TPCs (GET) AGET chips 64 data channels + 4 FPN channels Fixed Pattern Noise (FPN) channels measure the baseline due to electronics Preamplification and Shaping AsAd (ASIC and ADC) boards 4 AGET chips per board Digitizes signal from the AGET CoBo (Concentration Board) Module Up to 4 AsAd per CoBo MuTAnT (Multiplicity, Trigger and Time) Manages the trigger for the connected CoBos E.C. Pollacco et al. NIMA 887 (2018) 81-93 Curtis Hunt, TPC2023 Workshop , Slide 7
Identifying Rare Isotope Beams Beams are separated in a magnetic fragment separator Puts a very tight cut on rigidity (??) ?? ? ?? Measure time of flight (ToF) of beam between two fixed points Gives the particle velocity (?) Measure energy loss through gas detector ( E) ? ? Lab not convinced it was possible with lead -> new detector system! Z: Element number A: Mass number q: Charge-state Simulated PID plot ? ?= 2.4 ? Curtis Hunt, TPC2023 Workshop , Slide 8
HEavy ISotope Tagger (HEIST) ToF ? Adam Anthony MSU Grad Student Curtis Hunt, TPC2023 Workshop , Slide 9
Experimental Setup PID system (HEIST) MCP and ion chamber Si and HPGe support system Degrader system Fe degrader and ion chamber Active target (AT-TPC) 4He gas target E* from 65 MeV down to 30 MeV Trigger on MCP2 and pad plane coincidence Curtis Hunt, TPC2023 Workshop , Slide 10
Finding Window/Pad Plane Time Identify pads showing energy deposition near window and pad plane Fit leading edge of pad traces Window: t0gives time electrons drifted from window reach pad plane Plane: t0 gives time fission fragment hits pad plane Window Pad Plane Pad Plane Fit equation: Response function from electronics A exp ? ? ?? sin Bergen Kendziorski MSU Undergraduate Window Fit equation: Empirical estimation for spread from diffusion ? ? ?? ? 3 ? ?? ? ?? +1 ? exp Curtis Hunt, TPC2023 Workshop , Slide 11
Sorting Fission From Beam Low gain region in center of TPC should omit beam events In practice, we still record orders of magnitude more beam than fission ~ 2-3% of events are fission events Algorithm was developed to tag as many unwanted events as possible Set of ~2500 hand labeled events was used for development Parameters that can be quickly & automatically calculated were found Polar angles of straight-line tracks, spread of e- hits recorded in x-y plane, etc. Events are either labeled fission, beam, or other if parameters are ambiguous Beam Fission Other? Curtis Hunt, TPC2023 Workshop , Slide 12
Sorting Fission From Beam: Results Events are sorted into beam , fission , and other categories. ~2-3% of all events are sorted into the other category ~15-30% of fission events are sorted into other Run: 206 Beam Rate: 3809 Gas: CF4 @ 150 Torr Other : 27 Fissions 111 Beams Run: 261 Beam Rate: 2136 Gas: CF4 @ 200 Torr Other : 25 Fissions 44 Beams Combining fission and other categories yield 100% fission event capture with ~30-40% events being beam events Joe Wieske MSU Grad Student Curtis Hunt, TPC2023 Workshop , Slide 13
Machine Learning Approach To Classifying Fission and Beam Events Use a machine learning model to classify these Other events. Worked with Michelle Kuchera and Raghu Ramanujan (Davidson College) Train the model on labeled Fission and Beam events from the algorithmic approach Apply the trained model to a small subset of hand labeled Other events for testing 81% Fission events in Other correctly identified ~94-97% Fission events captured with algorithmic + ML methods combined Fission Beam Guess Beam Guess Fission True Beam 902 0 True Fission 6 25 Poulomi Dey MSU Undergraduate Curtis Hunt, TPC2023 Workshop , Slide 14
Space Charge: The Problem Accumulated positive ions (He+) drift much slower than e- because of their significantly lower mobility Creates a line of positive charge along the beam path Some fission fragment tracks curve inward Drifting e- are pulled inwards by the resulting ion transverse E-field Effect is rate dependent Results in wider folding angles and reaction vertices offset downstream Vertex location gives the excitation energy Fragment directions give mass split Curtis Hunt, TPC2023 Workshop , Slide 15
Space Charge: How To Correct Distortions Fission events are fit with Y-track to extract observables (vertex, angles) For certain events we can determine the vertex separately from Y-track The lighter fission fragments loose much less energy over distance than the heavier beam, creating a discontinuity in dE/dx These provide a metric to quantify how well corrections are tuned Proper correction will result in vertex agreement between the two methods ***Before correction*** Beam region dE/dx Curtis Hunt, TPC2023 Workshop , Slide 16
Space Charge: How To Correct Distortions Distortions to e- drift path can be modeled by solving Langevin Equation Electric field is a combination of the known TPC drift field and the unknown space charge distribution Space charge should build up more strongly towards entrance window Input a model for the space charge distribution, then solve for drift corrections Curtis Hunt, TPC2023 Workshop , Slide 17
Space Charge: Correction Results Vertex Difference - Before - Vertex Difference - After - ***Example Corrected Event*** ADC Vertex (mm) RED = Before correction BLUE = After correction Y-Fit Vertex (mm) Y-Fit Vertex (mm) Joe Wieske MSU Grad Student Curtis Hunt, TPC2023 Workshop , Slide 18
Deconvolution Shaper Effects Input Charge Output Signal When the charge distribution is short relative to the shaper rise time, the total charge is measured by the shaped pulse height When the distribution is long the pulse height does not represent the total charge Fission fragments move toward the pad plane, spending notable distance over a given pad Curtis Hunt, TPC2023 Workshop , Slide 19
Deconvolution Reconstruction Principle Output signal, sout(t), is the convolution of input signal, iin(t), and the response function, h(t) Channel dependent The signal can be deconvolved with Fourier transform ????(?) ?(?) ???? = With the Fourier transform of the response function the input charge as a function of time can be found Need response function Response function can be obtained from output with known input ????(?) ???(?) ????? = J. Giovinazzo et al. NIMA 840 (2016) 15-27 Curtis Hunt, TPC2023 Workshop , Slide 20
Deconvolution Response Function Shaped Pulser Signal [sout(t)] Pulser Signal [iin(t)] 650 mV internal AsAd pulser test used to get response function Pulser signal is recorded by the FPN channels Deconvolution of the pulser signal did not work due to very few time buckets involved in the signal Pulser signal is effectively a delta function Shaped pulser signal was used as the response function Curtis Hunt, TPC2023 Workshop , Slide 21
Deconvolution Response Function Shaped Pulser Signal [sout(t)] Response Function [h(t)] The inverted signal messes up the return to zero Add the positive signal to the negative signal to cancel it out Some glitches needed to be removed and extrapolated Shifted response function to start at 0 and normalized height to 1 Curtis Hunt, TPC2023 Workshop , Slide 22
Deconvolution Results Shaped Signal [sout(t)] Reconstructed Charge [iin(t)] Deconviolution corrects shape for undershoot due to response function Gives reliable dQ/dt results for fission fragments Curtis Hunt, TPC2023 Workshop , Slide 23
Preliminary Results Charge asymmetry Energy loss of products ?? ?? ? ?? Energy Loss related to mass of particles Adam Anthony MSU Grad Student ?? ?2 ?2?(?), where ?(?) varies slowly - Fragment A - Fragment B Charge Missing pads Time Buckets Curtis Hunt, TPC2023 Workshop , Slide 24
Preliminary Results Charge asymmetry Energy loss of products ?? ?? ? ?? Energy Loss related to mass of particles Adam Anthony MSU Grad Student ?? ?2 ?2?(?), where ?(?) varies slowly - Fragment A - Fragment B Charge Missing pads Time Buckets Curtis Hunt, TPC2023 Workshop , Slide 25
Summary Fission in AT-TPC Experiment looking at fission near 198Pb has been performed with AT-TPC Challenges Solved Beam particle identification HEavy ISotope Tagger (HEIST) Identifying Fission Events Use a combination of algorithmic and machine learning methods Space Charge Solved Longevin equation to correct fission tracks Signal Processing Deconvolution for accurate dQ/dt (related to dE/dx) Preliminary fission fragment asymmetry measurements More results, including cross sections, to come Future Use rare isotope beams from FRIB to look at fission of more exotic isotopes Curtis Hunt, TPC2023 Workshop , Slide 26
Questions? Questions? Curtis Hunt, TPC2023 Workshop , Slide 27
Backup Slides Curtis Hunt, TPC2023 Workshop , Slide 28
Sorting Fission From Beam: Example Parameter Curtis Hunt, TPC2023 Workshop , Slide 29
Sorting Fission from Beam: Results Other = 27 Fissions 111 Beams Run: 211 Beam Rate: 6467 Gas: CF4 @ 150 Torr Trigger: DS Singles + coinc Run: 210 Beam Rate: 3335 Gas: CF4 @ 150 Torr Trigger: DS Singles + coinc Run: 206 Beam Rate: 3809 Gas: CF4 @ 150 Torr Trigger: DS Singles + coinc Run: 271 Beam Rate: 7665 Gas: CF4 @ 200 Torr Trigger: DS Singles (500) + coinc Other = 25 Fissions 44 Beams Run: 261 Beam Rate: 2136 Gas: CF4 @ 200 Torr Trigger: DS Singles (500) + coinc Run: 260 Beam Rate: 2106 Gas: CF4 @ 200 Torr Trigger: DS Singles (5000) + coinc , Slide 30 Curtis Hunt, TPC2023 Workshop
Deconvolution Reconstruction Principle Output signal, sout(t), is the convolution of input signal, iin(t), and the response function, h(t) Channel dependent The signal can be deconvolved with Fourier transform ???? =????(?) ?(?) With the Fourier transform of the response function the input charge as a function of time can be found Analytical formula for the response function ? = ? exp 3? A is the amplification gain is the peaking time Analytical formula is not accurate enough for proper deconvolution 3 ? ? ? ? sin ? Curtis Hunt, TPC2023 Workshop , Slide 31
Deconvolution Baseline Corrections Pulser tests with internal AsAd pulser were used to obtain response function of the GET preamp and shaper 0 V test to get backgrounds Channel dependent baseline Phase effect from circular buffer Baselines usually corrected by FPN channels, but in pulser test FPN channels have the pulser signal Curtis Hunt, TPC2023 Workshop , Slide 32
Iterative Deconvolution For long signals (such as for the beam) the deconvolution does not reconstruct charge perfectly Curtis Hunt, TPC2023 Workshop , Slide 33
Iterative Deconvolution Signal can be reconstructed by convoluting the charge with the response function using the summation method The difference between the source signal and the reconstructed signal is not 0 Curtis Hunt, TPC2023 Workshop , Slide 34
Iterative Deconvolution The difference between the actual signal and the signal from the reconstructed charge can be iteratively used to correct the charge. 1) Use the deconvolution method on the signal difference to obtain the difference in charge space (go from signal space to charge space ) 2) Subtract the result from the charge and obtain a new reconstructed signal 3) Calculate the difference between the source signal and new reconstructed signal 4) Repeat Curtis Hunt, TPC2023 Workshop , Slide 35
Iterative Deconvolution 4 iterations appears to be sufficient to obtain a relatively flat difference Curtis Hunt, TPC2023 Workshop , Slide 36
Observables from AT-TPC Cross-section Energy determined from vertex location Space Charge corrections crucial Determination of Fission vs Beam Standard + ML methods Curtis Hunt, TPC2023 Workshop , Slide 37
Observables from AT-TPC Cross-section Energy determined from vertex location Space Charge corrections crucial Determination of Fission vs Beam Standard + ML methods Charge asymmetry Energy loss of products ?? ?? ? ?? Deconvolution important for charge comparisons ?? ?2 ?2?(?), where ?(?) varies slowly Curtis Hunt, TPC2023 Workshop , Slide 38
Simulation of fission events Experimental fission events (all beam species) Simulated fission events Z ~ 32 Z ~ 52 ?? ?? \ ?? ?? ?? ?? \ ?? ?? Curtis Hunt, TPC2023 Workshop , Slide 39
